(1) Field of the Invention
This invention relates to heterofunctional peptide conjugates derived, in part, from the p17 gag protein of human immunodeficiency virus (HIV) which is the causative organism of the disease known as Acquired Immune Deficiency Syndrome (AIDS). More particularly, the invention relates to the use of certain p17 peptide fragments and analogues as a component of a heterofunctional conjugate with a T cell binding ligand (TCBL) and to an AIDS vaccine based on such heterofunctional conjugate and to the use of the resulting AIDS vaccine compositions in a method for treating or preventing AIDS which method is based on the discovery that the heterofunctional peptide conjugates are capable of eliciting TH1 associated antibodies and other aspects of a TH1 cellular immune response.
(2) Discussion of the Prior Art
There has been extensive research over the past several years, first to identify the cause of AIDS and after the positive identification of the retroviruses generically referred to as HIV, as the causative organism, efforts have concentrated on more detailed analysis of the genetic makeup, molecular biology, pathogenesis, biochemistry, development of highly sensitive methods of detection of virus and antibodies and treatment protocols, and therapies. Extensive progress has been made in all of these areas yet much work needs to be done to effectively combat the spread of AIDS. An essential part of the approach for combatting the spread of the highly infectious HIV virus is the development of effective vaccines that stimulate the appropriate components of the immune system. In this regard, knowledge of the natural history of infection with HIV and the various immune responses and the clinical condition and the outcome may be useful in preparing effective vaccine preparations for either therapeutic or preventive means. In this regard accurate diagnosis of the stage of disease and the immune status with regard to HIV and the knowledge that certain means are available may encourage infected patients to alter or modify their lifestyles in such manner as to reduce the risk of spreading the virus.
Additionally, identification of protective antibodies based on epitope recognition can offer a more effective mechanism for staging and/or diagnosing the AIDS or pre-AIDS disease. Such staging and early diagnosis of seropositive individuals may then allow for vaccinations to provide the appropriate protective immune responses for treating the seropositive individual.
For many infectious diseases it is now recognized that the presence or absence of antibody is not as critical in the satisfactory clearance of the agent and disease resolution as much as it is the class or subclass of antibody, other immune system activities such as cytotoxic T lymphocytes and the particular antigens of the agent being recognized. For example, resolution of intracellular infections, such as tuberculosis, require activation of delayed-type hypersensitivity (DTH) and helper T cell activities, properties of CD4+ T cells (Finkelman et al. 1990, Ann Rev Immunol 8:303). CD4+ T helper cells have been delineated into TH1 and TH2 subsets in the mouse based on the cytokines produced and other activities which are promoted or associated with the T cell (Snapper et al 1987, Science 236:944; Salgame et al 1991, Science 254:279; Romagnani 1992, Immunol Today 13:379). In the mouse TH1 cells promote DTH, produce IL-2, interferon-.gamma. and promote IgG2a synthesis and inhibit TH2 responses, whereas TH2 cells produce IL-4, IL-10, promote B cell growth and differentiation leading to IgG1 and IgE production and inhibit TH1 responses (Bretsher et al 1992, Science 257:539). CD8+ cytotoxic T cell responses are also associated with TH1 responses (Cox et al.1992 Immunol Today 13:445). Based upon this information the nature of the T cell response generated by a specific immunization protocol can be evaluated by measurement of the specific lymphokine and/or antibodies produced (Finkelman et al. 1990, Annu Rev Immunol 8:303; Mosmann et al. 1989, Ann Rev Immunol 7:145). More recently it has been recognized that CD8+ cells responsible for CTL activity form a part of the cellular immune response mediated by TH1 cells (Taylor-Robinson et al. 1993, Science 260:1931).
HIV has several major classes of proteins referred to as the outer envelope structural group (env gp120 and p41), gag (or internal structural proteins p24, p17) and several nonstructural and regulatory genes and encoded proteins. Examination of the immune response and the disease state can be of assistance in designing new agents as vaccines. In one study, the response to HIV gag was polyisotypic (all/most classes or subclasses, IgM, IgG1, IgG3 and IgA) but antibodies to env were strikingly restricted to IgG1, (Khalife et al 1988, AIDS Res. and Hu. Retroviruses 4:3). The immune response to the nonstructural protein F (3'orf) also was restricted to IgM, IgA and IgG1 but not as much as the response to env and Khalife et al, ibid, further noted that IgG4 and IgE response which was gag restricted was found to be confined to the Hemophiliacs group.
Interestingly, in regard to man and HIV it is recognized that the disappearance of a particular subclass of antibodies (IgG3) to a conserved p17 protein molecule is most closely associated with disease progression. Antibodies to HIV of IgG1 and IgG2 subclasses are found in nearly all sera from HIV infected individuals at different stages of the disease. However, the presence of IgG3 is largely associated with gag proteins (p17, p24 and p55). Conversely almost all antibodies to the major viral surface proteins p41 and gp120 are of the IgG1 subclass and are present even in late stage disease (McDougal et al 1987 J. Clin, Invest 80:316-24).
The findings of McDougal et al ibid and also Jiang et al (J AIDS 1992, 5:382) are, in part, at odds with those of Broliden et al (1989 Clin. Expt. Immunol. 76: 216). Broliden ibid found IgG3 antibodies, predominantly against gag proteins at all stages of disease, and IgG1 which was against all HIV antigens declining in latter stages of the disease in contrast to McDougal et al ibid. Jiang et al ibid showed that anti p17 antibodies declined with disease progression and, in part, this was associated with antibodies to certain peptides of p17 including HGP-30. Broliden et al, ibid, refer to neutralizing antibody as being of the IgG3 subclass but conclude that no HIV neutralizing antibodies of IgG3 have been found. Broliden et al, ibid, in their study relates with HIV ADCC to IgG1 and not IgG3, (see also Ljunggren et al, 1987, 8, J.Immunol Meth 104:7; Clin Exp. Immunol, 1987, 73:343). Others suggest that ADCC is associated with IgG3. Often it is found that ADCC is associated with IgG3 ("The Human IgG Subclasses: Molecular Analysis of Structure, function and Regulation" Shakib, F. ed., Pergamon Press, 1990: Jeffries et al, Chapter 6, p.93; Pound et al, Chapter 7, p.111; Weiner, Chapter 8, p.135).
Rook et al (J. Immunol 1987, 138:1064) have shown that ADCC was associated with higher levels of antibodies to gag p24 than env. These workers did not examine antibodies to p17, various synthetic peptides of the proteins or the isotypes of the antibodies. Rook et al, ibid, also demonstrated higher levels of ADCC in asymptomatic HIV seropositive individuals than in AIDS patients.
The SCID Hu PBL mouse provides a very useful model that can be reconstituted with human cells and that can be infected with multiple strains of HIV (Mosier et al 1991, Science 251:791). This model allows for studying not only infection but also immunological intervention and control. Use of recombinant (r) gp160 (rgp160) for vaccination and demonstration of limited and short lived protection in the SCID model has been shown by Mosier et al (1993, Proc. Nat Acad. Sci 90:2443-2447). They showed the usefulness of this model in a preliminary report (Mosier et al, 1992, AIDS Res. Hu.
Retroviruses 8:1387). Anti HIV neutralizing antibody has been shown by passive immunization in this SCID hu model to afford some protection (Parren et al 1995, AIDS 9(6):F1-6).
An interesting finding for p17, one of the two major gag proteins, involves the 30 amino acid peptide HGP-30, whose sequence contains T and B cell epitopes immunoreactive with p17 of HIV (Sarin et al 1986, Science 232:1135; Naylor et al 1987, Proc Nat. Acad Sci 84:2951-5). See also U.S. Pat. No. 4,983,387 to Goldstein, et al, the entire disclosure of which is incorporated herein by reference thereto. This peptide of p17 known as HGP-30 has the following sequence:
Y S V H Q R I D V K D T K E A (SEQ ID NO:1) L E K I E E E Q N K S K K K A
HGP-30 has been conjugated to a large protein, Keyhole Limpet Haemocyanin (KLH), and found to be immunogenic in various animals and man, and the conjugate is well tolerated in both animals and humans (Gazzard et al, 1992, Vaccine Res. 1:129; Sarin et al, 1994, Vaccine Res. 3:495; Kahn 1992, AIDS Res Hu Retroviruses 8:1321; Naylor et al 1991, Int J. Immunopharm 13(Suppl):117). A pilot study of HGP-30 vaccine has shown protection from HIV infection in such SCID hu mice given PBL from HGP-30 immunized donors (Sarin et al 1995, Cell Molec Biol 41:401). More recently it has been shown that the presence of a predominance of IgG3 antibodies in serum of HGP-30 vaccine immunized human donors correlates with protection by PBL in the SCID Hu mouse HIV virus challenge model (Kahn et al 1996, Abstract 13 International AIDS Conference, Vancouver, Canada, July 1996; Talmadge et al 1996, Clin Immunol, Meeting New Orleans La. June 1996). This is thought to be quite different than what would be found if SCID hu mice with PBL from hyperimmunized env gp120 donors who had high titers of antibodies to the envelope protein where similarly infected with HIV; in this case it is expected that there would be afforded little or no protection.
With the recent recognition of the need to specifically direct the immune response various methods are under investigation. Use of protein carriers to direct nature of the immune response such as antibody, TH1, TH2, TS subclass of antibodies and CTL is now being characterized. Conjugation of HIV to Brucella abortus has been used to stimulate IgG2a in mice with the goal of stimulating IgG3 in humans, as shown by the following quote from the authors:
". . . From a functional point of view, IgG3 is the human counterpart of murine IgG2a . . . "
Golding et al, 1991, AIDS Res HU Retroviruses 7:435. This is based upon the evidence part of which was previously cited, namely, Pound et al, ibid, Weiner et al, ibid, Jeffries et al, ibid.
Considering the correlation of human IgG3 to murine IgG2a suggested by Golding et al, 1991, ibid the following considerations are also relevant. Protein antigens induce IgG1, IgG3 and IgG4 in man and IgG1 and IgG2 in mice (Hammarstrom et al, 1990, Chapt 15, p.301, in Shakib ed., ibid). Allergens induce IgG4 and IgE in man (and IgG1 and IgE in mice (Snapper, 1990, in Shakib ed., ibid). Carbohydrates induce IgG3 in mice and IgG2 in man (Hammarstrom et al, ibid). ADCC are associated with IgG1 and IgG3 in man and IgG2 in mice (Pound et al, 1990, in Shakib ed., ibid). Complement binding activities are associated with IgG1 and IgG3 for man and IgG2a and IgG2b in mice (Hammarstom et al ibid; Pound et al, ibid). TH1 cytokines such as interferon .gamma. inhibit IgG1 in man (Kawano et al, 1994, J. Immunol. 153:4948) and IgG1, IgG2a, and IgE in mice, and IgG2a in mice (Snapper, ibid). IL-4 and IL-5 which stimulate TH2 responses stimulate IgG1 and IgE in mice, and IgG1 and IgE in man (Lundgren et al 1989, Eur J Immunol 19:1311; Snapper et al, ibid).
The use of adjuvants to direct immune response to a cellular or humoral pathway has shown that in the mouse alum directs the system towards IgG1 production and a poor DTH response whereas Freund's complete adjuvant induces a slightly stronger DTH and IgG2a response (Hadjipetrou-Kourounakis et al, 1984, Scan J. Immunol 19:219; Kenney et al, 1989, J. Immunol. Meth 121:157-66; van de Wihgert et al, 1991, Infect Immun 59:2750; Golding et al, 1994, Am J. Trop. Med Hyg. 50(4) (suppl):33)). The route of administration for TH1 or TH2 directed by adjuvants from Syntex SAF-1 and RIBI has also been reported (Golding et al, 1994, ibid). Likewise, the use of cytokines in vaccine formulations has also been evaluated and although showing promising results such use of cytokines in vaccines are likely to be limited (Lachman et al, 1995, AIDS Res Hu Retroviruses 11(8):921; Bui et al, 1994, J. AIDS 7:799).
Traditionally, small peptides must be attached to carrier proteins in order to elicit immune responses. Often a large protein such as KLH is used. However, it has been observed that heterogenous (impure) KLH yielded a better immune response than did a more homogenous preparation (Herscowitz et al, 1972, Immunol. 22:51). Consider also that in man KLH induces IgG1 and IgG2 in early stages and also IgG4 after prolonged immunization (Bird et al, 1990, Immunology 69:355-360). Similiar observations with KLH conjugates were reported by Zimmerman et al, 1996b ibid. This is in spite of the fact that there is a need for a directed response with IgG3 as a marker (McDougal et al 1989 ibid; Kahn et al, 1996, ibid). It is desirable to find other methods to direct the response primarily or substantially to murine IgG2a or human IgG3, a TH1 associated pathway. The present inventors recognized that the carrier should not be directing the response in an undesired direction and since the KLH molecule seems to be predominantly directing the response in the TH2 direction it was concluded that another carrier should be considered. Likewise, other factors such as costs, ease of manufacture, and stability, helped lead to the discovery of this invention.
There is no evidence that changing the nature of a peptide, such as by addition(s) or deletion(s) of one or several amino acids and/or method of attachment to a carrier could or would influence the subclass of antibody generated that recognize the epitope even though it is known that such manipulations can induce different responses such as the stimulation of B and T cells, cytotoxic and lymphoproliferative responses. However, the effect of T cell epitopes on antibody responses that has been reported has been for the presence or absence of antibody production by helper or suppressor immune responses (Sercarz et al, 1991, Immunol Today 12:111; Yowell et al, 1979, Nature 279:70; Wicker et al, 1984, Eur J. Immunol. 14:442).
Zimmerman et al (1996a,b, Vaccine Res 5:91-102, 103-118; WO 89/12458, the disclosures of which are incorporated herein in their entireties, by reference thereto) have taught that addition of a TCBL to a peptide epitope could alter the nature of the immune response (i.e., TH1 or TH2). They further showed that the antibodies derived from use of heteroconjugates were better able to recognize the native molecule than were the antibodies prepared by using a peptide-KLH conjugate. It was shown that the antibodies induced by the heteroconjugate had a broader specificity, so that they recognized the peptide epitope not only in the linear form, but also in the native molecule which should offer a number of advantages. Indeed, it was shown that the antibodies in some cases generated as a result of the use of the peptide conjugated to KLH were not able to recognize the epitope in the native molecule. Where comparisons exist this contrasts with the results of Vordermeier et al (1995, Vacc 13:1576). They showed that antibodies raised to the peptide epitope or to the recombinant protein 38 KD expressed in a bacterial system. This 38 KD protein (with the primary mycobacterium amino acid sequence, also had a leader sequence and was not processed as was the native protein in the Mycobacterium infected cell, such as, addition of carbohydrates, lipids and phosphorylated or sulfonated, were not able to recognize the native protein and were not protective upon challenge.
Prior work has established that p17 can be subdivided into several peptides for induction of immune responses (Goldstein et al, U.S. Pat. No. 4,983,387; Sarin et al, EP 0 246 829; Jiang et al, 1992, J. AIDS 5:382). That and other work has shown that in the p17 molecule numerous immunological epitopes are predicted to be present or experimentally determined (Naylor et al, 1990, Mono. Viro 18:74; Coates et al, 1987, Nature 326:549; Wahren et al, 1989, J AIDS 4:448; Broliden et al, ibid; Papsidero et al, 1989, J. Virol 63:267; Boucher et al, 1990, Clin Lab Anal 4:43). However, these reports studied total antibody response, cytotoxic T cell or lymphoproliferative responses. Other work has shown that certain conjugates which include a T cell binding ligand (TCBL) and an epitope of interest from a disease associated antigen can also have biological activity even when the epitope alone is inactive (Zimmerman et al 1996a,b, ibid). Since in man KLH induces IgG1 and IgG2 in early stages and also IgG4 after prolonged immunization (Bird et al, ibid) it is desirable to use other methods to direct the response to IgG3, a TH1 associated pathway, since the need for a directed response with IgG3 as a marker is well documented (McDougal et al, 1988, ibid; Kahn et al 1996, ibid).